ESMO 2010

Professor Alberto Sobrero - San Martino Hospital, Genoa, Italy

Clinical trials evaluating treatment of colorectal cancer

Interviewer, Gordon McVie. 

GM: Professor Sobrero, you’re the big Italian expert on colorectal cancer, European expert maybe, from Genoa. Thanks for giving us a couple of minutes. Now the sessions in colorectal cancer here at ESMO?

AS: Well we just finished that session and that was devoted to four major trials, three were phase III trials and they were devoted to two new agents and one agent that is already on the market. The first two presentations dealt with Cediranib and there was very much hope because of the rationale that supported this drug. It is a more complete inhibitor of VEGF TK so theoretically it should have worked. Instead, both trials were negative. By negative I don’t mean that they were negative under every respect, actually the first trial, the HORIZON 2 was positive in terms of PFS but you have to take into consideration that the first trial compared just chemotherapy versus chemotherapy plus Cediranib. So you have to compare with the standard of care today, that is chemotherapy plus Bevacizumab, and that’s the content, the goal, of the second trial, HORIZON 3 and that is where the drug actually failed because it was no better, actually. Quality of life was worse, treatment duration was shorter and there was no better response rate, PFS or survival. So we were so disheartened about that.

GM: But we have to do those studies and it’s good to get the negative data out there and people don’t need to re-do the study and they can take some lessons forward. My view is that you frequently, given the evidence of resistance in the face of a mechanism that should have worked, and you go back and sometimes you can find out what the cause of the resistance is.

AS: Yes, you are absolutely right but the issue here is did we really need more than 2,000 patients to answer that question? That leads us to the second part of the session and that was the paper dedicated to this new agent, the hedgehog inhibitor in advanced colorectal cancer. Again this was presented by Jordan Berlin and was a 200 patient study that was chemotherapy Bev plus/minus the new hedgehog inhibitor, isn’t that a wonderful design? 200 patients, so for a randomised phase II that should give a clear-cut answer whether to go or not go with a 2,000 patient study. And it gave a clear-cut answer – no benefit.

GM: Red light.

AS: Red light. Even Jordan Berlin, the presenter, concluded very nicely and correctly that that was a sign not to go ahead. So here we are, we’re left with the last presentation and the last presentation did not deal with novel agents, it dealt with an agent that is on the market and that is Cetuximab. Now the Nordic countries ran a very nice study of their regimen of FOLFOX, their variation, plus or minus Cetuximab. Now the disheartening part here is that no benefit was found.

GM: Why, do you think?

AS: Well I had to discuss the paper and I crushed my head over that. You obviously always have three classical interpretations – first the investigators were biased, so bias, and that was not the case.

GM: Unlikely.

AS: Well it’s a no. This group of investigators did not present this study at ASCO, they decided to postpone it because they wanted to be sure of the results, that’s amazing. So bias is out. Chance? Yes, this is the only study showing a lower response rate in the KRAS wild-type to the combination of chemo plus Cetuximab than in the KRAS mutated, one out of eleven. So you say that’s an outlier. Well, if you look at the confidence interval, chance may still play a role but you get the feeling that that’s extremely unlikely so you have to accept the third possible interpretation and that is true interaction, with what? With Oxaliplatin or with the schedule of a few. Now it’s more difficult to accept the notion that the schedule of a few, same drug, different schedule, induces a negative interaction, whereas if you look at the whole story of FOLFIRI plus Cetuximab, there is a solid, rocky, granitic story. And the story of Oxaliplatin, FOLFOX, plus Cetuximab – more shaky. Overall I presented four or five pieces of evidence for this type of interpretation. So with all the caution we can conclude that the message is that Oxaliplatin is not a good partner for Cetuximab overall whenever it is used in first line.

GM: I missed your discussion but that summary was brilliant.